The Extended Mind: Born to Be Wild? a Lesson from Action-Understanding
Total Page:16
File Type:pdf, Size:1020Kb
Phenom Cogn Sci (2011) 10:377–397 DOI 10.1007/s11097-011-9198-y The extended mind: born to be wild? A lesson from action-understanding Nivedita Gangopadhyay Published online: 12 February 2011 # Springer Science+Business Media B.V. 2011 Abstract The extended mind hypothesis (Clark and Chalmers in Analysis 58(1):7– 19, 1998; Clark 2008) is an influential hypothesis in philosophy of mind and cognitive science. I argue that the extended mind hypothesis is born to be wild. It has undeniable and irrepressible tendencies of flouting grounding assumptions of the traditional information-processing paradigm. I present case-studies from social cognition which not only support the extended mind proposal but also bring out its inherent wildness. In particular, I focus on cases of action-understanding and discuss the role of embodied intentionality in the extended mind project. I discuss two theories of action-understanding for exploring the support for the extended mind hypothesis in embodied intersubjective interaction, namely, simulation theory and a non-simulationist perceptual account. I argue that, if the extended mind adopts a simulation theory of action-understanding, it rejects representationalism. If it adopts a non-simulationist perceptual account of action-understanding, it rejects the classical sandwich view of the mind. Keywords Extended mind . Action-understanding . Simulation theory. Non- simulationist perceptual theory. Embodied intersubjectivity . Representationalism . Dynamical systems . Perception . Action . Social cognition Introduction The extended mind hypothesis has officially been in town for more than a decade. Clark’s Supersizing the Mind (2008) marks 10 years since the publication of the extended mind proposal by Clark and Chalmers (1998). The view that the mind oozes out of the boundaries of the brain to encompass the body and the environment of the situated agent has been described under various names, e.g. vehicle N. Gangopadhyay (*) Center for Subjectivity Research, University of Copenhagen, Njalsgade 140-142, 2300 Copenhagen S, Denmark e-mail: [email protected] 378 N. Gangopadhyay externalism, active externalism, how externalism and enabling externalism (Hurley 1998, 2010).1 In this paper, “vehicle externalism” and “extended mind” are used interchangeably. The original formulation of the extended mind hypothesis defends it in the context of a particular type of mental states, namely, beliefs. Clark and Chalmers spell out the proposal as follows, “…we will argue that beliefs can be constituted partly by features of the environment, when those features play the right sort of role in driving cognitive processes. If so, the mind extends into the world.” (Clark and Chalmers 1998, 12). The primary motivation for the extended mind hypothesis comes from three sources: (1) thought experiments, (2) everyday use of artefacts and (3) promising application of the extended cognition hypothesis in the domains of robotics and artificial intelligence. In this paper, I argue that the extended mind hypothesis is born to be wild. It has undeniable and irrepressible tendencies of flouting certain grounding assumptions of the traditional information-processing or computationalist/representationalist para- digm in philosophy of mind and cognitive science, the paradigm which took over the field following the publication of David Marr’s Vision (1982). I claim that going radical in some way is inevitable for the development of the extended mind approach. To defend the claim case-studies from social cognition are presented which not only support the extended mind proposal but also do so only by bringing out the proposal’s inherent wildness. In particular, the paper focuses on cases of action-understanding and discusses the role of embodied intentionality2 in the extended mind project. It explores the embodied engagement between intentional agents in search of an account of action-understanding which grounds the extended mind proposal. It discusses two theories of action-understanding as potential candidates for exploring the extent to which the extended mind hypothesis finds support in cases of embodied intersubjective interaction. These are: (1) the simulation theory (Gallese et al. 1996; Goldman 2006) and (2) a non-simulationist perceptual account of action-understanding (Gallagher 2008; Gallagher and Zahavi 2008).3 I argue that no matter which theory of action-understanding the extended mind hypothesis adopts to marshal the support it finds in cases of embodied 1 The extended mind hypothesis I shall consider in this paper is not concerned with providing a definition of the mind or a mark of the mental. I shall follow Clark’s(2008) and Chalmers’ (2008) pragmatic insight that where one sets the boundaries of the mind is largely a matter of what one wants to explain. This paper is also not concerned with metaphysical views which challenge the division between the mind and the world and stress their co-emergence (Thompson 2007; Zahavi 2008). I shall follow the original hypothesis of the extended mind theorists (Clark and Chalmers 1998) that some cases of ongoing agent–environment interaction qualify as cognitive processes in their own right even if they are not conducted entirely “in the head” and thus an account of mental states need not be radically neurocentric. 2 “Embodied intentionality” refers to the kind of intentionality that characterises our skilful bodily coping with the world. Much contemporary discussion of embodied intentionality is inspired by Merleau-Ponty’s (2002) idea of “motor intentionality”. He writes, “...my body appears to me as an attitude directed towards a certain existing or possible task.” (p.114). Again, “…the recognition of something between movement as a third person process and thought as a representation of movement- something which is an anticipation of, or arrival at, the objective and is ensured by the body itself as a motor power, a “motor project”…a “motor intentionality”….”(pp.126-127). 3 Theory–theory approaches to mind-reading (Gopnik and Wellman 1994; Baron-Cohen 1995) fall outside the scope of the present paper because they primarily focus on how we attribute propositional attitudes to others rather than how we understand the other’s embodied intentionality. The extended mind: born to be wild? 379 intersubjectivity, the hypothesis spills over fundamental traditional assumptions of the information-processing paradigm. If the extended mind hypothesis adopts the simulation theory to successfully ground itself in cases of embodied intersubjectivity, then the hypothesis is no longer confined within a representationalist framework. If the extended mind hypothesis opts for the non-simulationist perceptual account of action-understanding, then the hypothesis challenges the “classical sandwich” model of the mind (Hurley 1998) as upheld by the traditional information-processing approach. I leave it up to the extended mind theorists to decide which of these two ways of going wild in the context of action-understanding they prefer, because not going wild is not an option for them.4 Before we begin, here is a brief sketch of the traditional assumptions which I claim are challenged by the extended mind hypothesis in this paper. Both assumptions are accepted by Clark and form the grounding principles of his philosophical works. If the extended mind hypothesis goes wild, it may be a good thing for the hypothesis itself but it will be particularly tricky to fit it in with the rest of Clark’s (mostly) well-behaved philosophical views. First, the traditional information-processing paradigm is built on the concept of internal/mental representations. The intricacy of the concept of mental representation far exceeds this simple statement about it. The notion of mental representations is the most discussed and applied conceptual tool for explaining mind and behaviour in traditional theories. A unanimous definition of the term “representation”, as it is used in philosophy of mind and cognitive science, is yet to emerge. The term has been variously interpreted as well as “updated” (Clark and Toribio 1994; Wheeler and Clark 1999) according to the demands of the theory where it is used. However, in general, appeal to the notion of mental representations for explaining the mind remains the distinguishing feature of traditional theories. Indeed, discarding the idea of mental representations is often the first step taken by non-orthodox views when they attempt to construct theories outside the traditional paradigm (Noë 2004, 2010; Chemero 2009; Garzón 2008). A recent general definition of mental representation offered by Dietrich and Markman (2003) somewhat captures the gist of the notion. They write, “…a representation is any internal state that mediates or plays a mediating role between a system’s inputs and outputs in virtue of that state’s semantic content. We define semantic content in terms of information causally responsible for the state, and in terms of the use to which that information is put.” (Dietrich and Markman 2003, 97). A theory stepping out of the traditional school of thought by rejecting the notion of mental representation rejects a notion broadly resembling the above-mentioned definition. Second, traditional information-processing approach upholds a strict separation between perception, action, and cognition or between sensory, motor, and cognitive processes. The traditional model of the mind where perception, action, and cognition proceed as virtually independently of each other is dubbed “the